تالیفات


عنوان: A New Adaptive Steganographic Method with Scalable
تاریخ انتشار: 2013
فایل: دانلود
توضیحات:
  1. Seyed Rahman Soleimani, S. Haghbin, Masood Niazi, A New Adaptive Steganographic Method with Scalable Capacity and High Vision Quality, Journal of Passive Defence Science & Technology ADST, VOL 4, NO 1, May 2013

Abstract: In the present paper, a new adaptive steganographic method has been proposed to embed secret data in an image. Proposed method allows the user to set algorithm parameters in a way that reducing the distortion resulted from stegnanography in the image in spite of embedding secret data in all pixels of image, considering size of secret data required for sending to destination. These parameters can be also used as a secret steganographic key. In this method, considering the human vision system, more secrete data bits are embedded in edged areas of image than in its smooth ones. In the proposed method, blocks are classified into different groups based on the extent of local complexity of pixels, and the capacity of each group is determined by total size of secret data. This method is also resistant to statistical and histogram steganalysis attack.

نوع: مقاله و ژورنال

عنوان: A New High Quality Vision Non-Adaptive Steganograp
تاریخ انتشار: 2013
فایل: دانلود
توضیحات:

Seyed Rahman Soleimani, Masood Niazi, A New High Quality Vision Non-Adaptive Steganographic Method, Using Module and Combined Functions, International Journal of Emerging Trends in Signal Processing, VOL 1 ,Issue 2, Jan 2013

Abstract— In this paper, a new non-adaptive steganographic method for embedding the secret data in images, has been proposed. In this method, cover image is divided into non-overlapping two-pixel blocks using module and combined functions, the fixed number of secret bits will be hidden in each block. More ever, in this method we have tried to increase the quality of the stego image by reducing the distortion of stego pixels. The results, also confirms that the quality of stego image has been increased in our proposed method compared the other non-adaptive steganographic methods.

نوع: مقاله و ژورنال

عنوان: Hirarchical Harmony Linear Discriminant Analysis
تاریخ انتشار: 2011
فایل: دانلود
توضیحات:

 

Abstract—Linear Discriminate Analysis is commonly used in feature reduction. Based on the analysis on the several limitations of traditional LDA, this paper makes an effort to propose a new way named Hierarchical Harmony Linear Discriminant Analysis (HH-LDA), which computes between class scatter matrixes optimally. It is reached by combining hierarchical scheme and Harmony Search (HS) algorithm. In this paper, a pre-processing step is proposed in order to increase accuracy of classification. The aim of this approach is finding a transformation matrix causes classes to be more discriminable by transforming data into the new space and
consequently, increases the classification accuracy. This transformation matrix is computed through two methods based on linear discrimination. In the first method, we use class dependent LDA to increase classification accuracy by finding a transformation that maximizes the between-class scatter and minimizes within-class scatter using a transformation matrix. Because LDA cannot obtain optimal transformation, in the second method, Harmony Search is used to increase performance of LDA. Obtained results show that
utilization of these pre-processing causes increasing the accuracy of different classifiers.

نوع: کنفرانس

عنوان: چهارچوبی براي تشخیص خاتمه پذیري قوانین فعال با است
تاریخ انتشار: 2011
فایل: دانلود
توضیحات:

نعمت اله کمال فر، مسعود نیازي ترشیز، مجید وفایی جهان، چهارچوبی براي تشخیص خاتمه پذیري قوانین فعال با استفاده از وارسی مدل مبتنی بر SAT و BDD، اولین کنفرانس ملی دانش پژوھان کامپیوتر و فناوری اطلاعات، 1390

چکیده: یکی از مهمترین مشکلات مربوط به قوانین فعال، مسئله خاتمه ناپذیري آنهاست. تاکنون روشهاي متعددي براي بررسی خاتمه پذیري قوانین فعال پیشنهاد شده اند. بسیاري از روشهاي پیشنهادي، محافظه کارانه عمل نموده و قادر به اظهار نظر قطعی در مورد خاتمه ناپذیري قوانین نمیباشند. در این میان چند روش نیز با بهره گیري از وارسی مدل، موفق به آنالیز خودکار و غیر محافظه کارانۀ خصوصیت خاتمه پذیري قوانین فعال گشته اند. اما این روشها نیز از مشکل اصلی اکثر روشهاي مبتنی بر وارسی مدل، یعنی انفجار فضاي حالت رنج میبرند. همچنین اکثر روشهاي پیشنهادي در گذشته، از استراتژيهاي مختلف پردازش قوانین پشتیبانی نمینمایند. بهره گرفته ایم. بر اساس SAT و BDD ما در روش  پیشنهادي خود به منظور اجتناب از مشکل انفجار فضاي حالت، از وارسی مدل مبتنی بر چهارچوب پیشنهادي ما، خصوصیت خاتمه پذیري قوانین فعال تحت استراتژيهاي مختلف پردازش قوانین و با استفاده از وارسي مدل قدرتمندي به نام  NuSMV2  قابل بررسی میباشد. بعلاوه، چهارچوب پیشنهادي به دلیل عدم وابستگی ساختاري به معماري پایگاه داده فعال، به راحتی براي تمامی سیستمهاي مبتنی بر قوانین فعال قابل تطبیق و استفاده است. 

نوع: کنفرانس

عنوان: FUMOR :FUzzy Multi-hOp Routing for Wireless sensor
تاریخ انتشار: 2011
فایل: دانلود
توضیحات:

Mohammad Mahdi Tajari, Mohammad Hosein Yaghmaii Moghaddam, Masood Niazi Torshiz, Javad Artin, FUMOR :FUzzy Multi-hOp Routing for Wireless sensor network, 

 

<div style="\&quot;direction:" ltr;\"="">

 

Abstract—In order to increase the life time of wireless sensor networks, using hierarchical methods which transfer its data toward sink through multi-hop would be suitable and significantly decreases the energy consumption, Also the management of networks with a lot of Nodes would be done much better. In this paper we have presented fuzzy hierarchy method FUMOR (Fuzzy Multi-hop Routing for Wireless sensor network) and the goal is to increase the life time of the network. The fuzzy logic which has a great control power is to control uncertain conditions used to choose gateway (cluster head). FUMOR method toward node data through multi-hop to increase network life time. In our proposed method we used factors like: Energy Parameters, relative Distance and Level of the Nodes as the inputs of fuzzy module. The stimulations have shown that the proposed idea has a good efficiency.

 

نوع: کنفرانس

عنوان: CFGA : Clustering wireless sensor network using fu
تاریخ انتشار: 2011
فایل: دانلود
توضیحات:

Abstract - Wireless sensor networks have numerous nodes with limited energy which have the ability to monitor around themselves and these nodes are scattered in a limited geographic area. One of the important issues in these networks is increasing the network lifetime. In this study, we introduce an efficient protocol for trade-off between loads and increase in life expectancy of the network, known as CFGA (Clustered wsn using fuzzy logic and genetic algorithm) which uses the single - step method for intracluster communication and multi - step for intercluster communication. At the beginning of each round, each node first checks its fuzzy module, and based on output of the fuzzy module, if the clusterhead capability exists for node, it would be ready ( in each region, number of the best nodes would be ready), then at the base station by using genetic algorithm and the location of clusterheads based on minimum energy consumed, the optimum network nodes are determined. In this study, consumption of energy for nodes which are not capable of becoming clustered is prevented and also only nodes with high capabilities in genetic algorithm take part in order to converge faster to the optimum solution. 

 

نوع: کنفرانس

عنوان: Mining Generalized Fuzzy Association Rules via Det
تاریخ انتشار:
فایل: دانلود
توضیحات:

Ehsan Vejdani Mahmoudi, Vahid Aghighi, Masood Niazi Torshiz, Mehrdad Jalali, Mahdi Yaghoobi, Mining Generalized Fuzzy Association Rules via Determining Minimum Supports

 

 

<div style="\&quot;direction:" ltr;\"="">

 

Abstract: Association rule mining is based on the assumption that users can specify the minimum-support for mining their databases. It has been identified that setting the minimum support is a difficult task to users. This can hamper the widespread applications of these algorithms. This paper proposes a method for computing minimum supports for each item. It therefore will run the fuzzy multi-level mining algorithm for extracting knowledge implicit in quantitative transactions, immediately. More specifically, our algorithms automatically generate actual minimum-supports according to users’ mining requirements. In order to address this need, the new approach can express tow profits includes computing the minimum support for each item regarding to characteristic for each item in database and making a system automation. We considered an algorithm that can cover the multiple level association rules under multiple item supports. We experimentally examine the algorithms using a dataset, and demonstrate that our algorithm fittingly approximates actual minimum-supports from the commonly-used requirements.

 

نوع: کنفرانس

عنوان: DESIGN AND PERFORMANCE EVALUATION OF A FUZZY-BASED
تاریخ انتشار: 2009
فایل: دانلود
توضیحات:

MASOOD NIAZI TORSHIZ, ALI MOVAGHAR, DESIGN AND PERFORMANCE EVALUATION OF A FUZZY-BASED TRAFFIC CONDITIONER FOR MOBILE AD HOC NETWORKS, Journal of Circuits, Systems, and Computers, Vol. 17, No. 6 (2008) 995–1014

 

Abstract: A mobile ad hoc network is a collection of mobile hosts forming a temporary network on the fly, without using any fixed infrastructure. Characteristics of mobile ad hoc networks such as lack of central coordination, mobility of hosts, dynamically varying network topology, and limited availability of resources make QoS provisioning very challenging in such networks.
In this paper, we introduce a fuzzy QoS traffic conditioner for mobile ad hoc networks. The proposed traffic conditioner consists of fuzzy admission control (FAC), fuzzy traffic rate controller (FTRC), and fuzzy scheduler (FS). The proposed FAC monitors the delay and available bandwidth and decides whether to accept or reject the request. The FTRC uses the additive increase multiplicative decrease (AIMD) rate control algorithm as a base, in which a node increments its transmission rate when the observed delay is below the predefined threshold, with an increment rate of c Kbps and decreases its transmission rate by r% when the delay passes the threshold. FTRC accepts the packet delay and the “delay-threshold d” as inputs and calculates c and r by using a set of fuzzy rules. The third part of the proposed traffic conditioner is FS, which is based on the traditional weighted round robin (WRR) mechanism. FS monitors the packet drop and delay of each queue and adjusts the queue weights by using a fuzzy inference engine.

نوع: مقاله و ژورنال